Announcement

Collapse
No announcement yet.

New Tribes' Populations

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • JuanCarlos
    replied
    "The inclusion of new reference data have allowed us to update our world region analysis."

    I wonder what they mean by this.

    Also, why do they have so many Brazilian populations?

    Leave a comment:


  • tomcat
    replied
    Thank you Gene for posting the new populations. I suppose I should be interested in the Swedish data for the sake of the unknown GF named Nelson.

    Leave a comment:


  • rainbow
    replied
    I just looked at my July 2009 DNATribes again and wrote down the matches that had a tribescore of (0.5) and higher. Nothing in the Native American panel. Nothing in my Europa. The main report, Part B: Syria (0.6) 183.82. Part C global/diaspora: Puerto Rican (0.88) 217.73, and Syria (0.6) 183.82, and Santa Catarina, Brazil (0.69) 118.34. Part D world region nothing. Extended has, in addition to the above, #59 Caucasian (Australia) (0.78) 52.92, and #165 Vienna, Austria (0.62) 21.20, and #246 Costa Rica (0.5) 11.67. And that's it.

    Leave a comment:


  • tomcat
    replied
    Tribes Scores are those parenthetical scores beside the population to which one is matched. They 'grade' your match score relative to the population as a whole. I only give credence to average Tribes Scores (0.50 range). A match that is Tribes scored at 0.90 means that your profile is unlike those of 90% of the population. Such a high score seems, to me, as dubiously valid as a very low Tribes score on a match.

    So Tribes Scores are only qualifiers on Match Scores. And the lower the score range for matches the greater the number of populations one will match. In example, on a Tribes Extended Report of more than a year ago, at a Match range of 4.0 - 5.0, I have equivalent matches in Asia Minor, Eastern Europe, the Mediterranean, the New World, North Africa and Northwest Europe. Not very useful. But in the 16.0 - 17.0 range I only have matches to Turkey and Bogota. And those are 'closer' to known ancestry in their equivalency.
    Last edited by tomcat; 21 September 2009, 09:05 PM.

    Leave a comment:


  • J. Fold
    replied
    dna tribes wide net

    If all tribes scores of .05 and above are within ball park range- than this is a really wide net of matches is it not? And leaves few poplulations that one does not match- and if one can match neighboring areas- as much as primary location of ancestry as Dna Tribes explains- than we can also match distant areas- that may simply mean that we have common ancestry with thousands of years ago- which leaves me with the notion that this test is not precise, but rather an incredible macro- matching excercise for the sake of large population connections- not micro matching appropriate for one individual..

    :

    Leave a comment:


  • rainbow
    replied
    Hi Gene, I am posting from my phone now and don't have access to my DNATribes now. It is saved to my Compaq, at home. I think I recall that my Pakistan match(es) have Tribescores around (0.34), which is in the average range. About double my Scottish tribescores, around (0.17). My mother is basically half Scottish and half English, including small amounts of Welsh and Alsatian(french german swiss area). She thinks DNATribes ridiculous. I learned to give up trying to discuss it with her. She does have hair and eyes that look sorta gypsy or Iraqi to me. I've told her she looks like she could be Saddam Husseins sister.

    Leave a comment:


  • rainbow
    replied
    Thanks Maria. I want to see the new populations/categories. And thanks Gene. It would be great if DNATribes label each line in the extended report. I used to go bonkers from that. I have and have had extended matches to the Amazon Region and Alaska and Javanese and Mozambique and Kathmandu, Nepal, and Bahrain, and so on, that have freaked me out. I hope they fix their algorithms some more (I still have extended matches to Pakistan and Maori). And I hope they label each as diaspora or native. The Amazon Region match was diaspora.

    Leave a comment:


  • Maria_W
    replied
    Sept 30....

    DNA Tribes has pushed their poplation update back to Sept 30!

    Maria

    Leave a comment:


  • tomcat
    replied
    I never have discussed Tribes' algorithms with Ann Turner.

    I think the standard matching algorithm, and Tribes version of it, work surprisingly well. But, if one is admixed, as most of us are, our admixed profiles drive our matches whereas we would like our several ancestry components to register on their own terms.

    In example, I am one-half Ukrainian Ashkenazi and one-half American Frontier Blend with a significant Native component. It ought to be possible, with today's computer and software technology, to run every possible combination of my alleles to yield a series of paired results; 1a and 1b and so forth, where #a is the mirror of #b, the alleles in #a are 'opposite' or other than the alleles in #b.

    In my case, I woud expect that relatively few of my potential haplotyes would register as 100% Ukrainian Ashkenazi and that their opposites would all register as Native, to a degree.

    Through a follow-on analysis of those pairs that best conform to known ancestry it may be possible to disclose minor ancestry components.

    This may not work in cases of multiple admixture.
    Last edited by tomcat; 9 September 2009, 10:38 PM.

    Leave a comment:


  • jah
    replied
    apology & query (via synthesis of ideas)

    Tom, it's been a long time since we last corresponded. I wanted to apologize if I was offensive in any previous exchange. Also, I think that your idea about possible Slavic slavery in Andalusia as a part of my ancestry seems very plausible. You may be interested to learn that in tracking my ancestral paper trail, I discovered that my Spanish mitochondrial ancestress is in fact real, and very recent (last 150-200 years).

    Anyway, have you ever discussed Tribes' algorithms with Ann Turner? (Pardon me if you have posted on this previously. If so, could you direct me accordingly?)

    Thanks, John

    Leave a comment:


  • tomcat
    replied
    One can hope. Maybe next year they will give up their dubious Deep Ancestry analsis and get practical. Improbable Crosses. Lesotho x Evenki=Michigan Soccer Mom.

    Leave a comment:


  • Maria_W
    replied
    Time to change algorithms...

    Tom,

    I agree, I always have to go to their website or email them..
    Yes, wish they would change their algorithms also...
    Will see what is new the middle of the month...

    Maria

    Leave a comment:


  • tomcat
    replied
    Thanks Maria.

    DNAT might promote these additions; "Coming in September 2009 ..."

    I don't expect to update unless the new populations are relevant. New populations are a given. Tribes ought to be working on new algorithms.

    Leave a comment:


  • Maria_W
    replied
    DNA Tribes...

    New population data comming in mid September per Lucas at DNA Tribes...

    Maria

    Leave a comment:


  • rainbow
    replied
    Originally posted by BlackWolf View Post
    Just got back my Native American panel, only one match

    Chol Maya 0.01 (0.01)
    Hi,

    That's your top match, meaning it is at the top of your list, but it isn't actually a match. A "match", according to DNATribes, is 1.00 and above.


    At the top of my list I have Athabaskan (Alaska) at 0.01 (0.01). In some updates it is under 1.00 and not a match. And in other updates it is over 1.00 and a match.

    Leave a comment:

Working...
X